ispcpphptemp

[kilburn]

Nevermind about the "IFS" thing, as it's a stupid fault on my side. Your bash behaviour is actually good, and there's an error on the ispcpphptmp.sh script that I've just fixed in trunk.

Now, if you run the script twice in a very short time, the second run is expected to end almost instantly because the file properties are already in the disk cache (left there by the previous execution). If you wait some minutes, they'll have been removed from the cache and will need to be read from disk again ("stat"ing a lot of files is the slow part of this script).

That said, I would need a trace of one of this "never ending" processes to actually see what's happening there (the trace you posted earlier is from an execution which finished correctly, right?).